What is the goal of dockex?
The goal of dockex is to create a community-driven resource portal with a twist. What if someone can perform a search for a resource and retrieve results that are best suited for THAT particular user? What if the system continues to learn about the user's interest and his willing to expand in a particular field (resource) and so continue to provide even better results? What if the system provides the capabilities to increase the value of a the resource-provider (user) and grow to become a highly valued member by the community?
The basic principle of dockex can be summed up in this example:
You have a sample work that you have done (be it a "linux driver", an "essay", or a "collection of pictures", etc...) and you share it with the community for feedback. Three people (A, B, C) value your work and each provide you with a score. Now, current rating and voting systems found in ebay.com, digg.com, etc... take a linear approach and gather the scores and ultimately give you the average of the votes, which then you consider as your final feedback. The problem is... if person A (e.g. Linus Torvalds) has more experience in that particular field wouldn't you value his input more than than B's, or C's? Of-course you would. And this where lies the problem in current community sites.
How does it compare with other community-driven sites?
Sites such as digg.com, slashdot.org, flickr.com, youtube.com, etc ... provide excellent resources for their respective users in general. However, they lack the user-focused resource and instead provide resources that the whole community thinks is appropriate. This works great sometimes, but not all the time.
dockex is not in a position to compete with any of the above sites. It's goal is not to become a news-community-driven-portal such as digg.com and slashdot.org, or a photo-community-driven-portal such as flickr.com, or video-community-driven-portal such as youtube.com, etc... instead it realizes the importance of these resources and works (with the help of the community) to drive such resources to the appropriate users depending on their interests.
How does it work?
Each user in dockex can submit an "item". An item is a link to a resource somewhere in the internet with a set of tags attached to it. Now, each of these tags is owned by the submitter. Starting from a clean-slate these tags are each of 50% (Neutral) value. Meaning, this user is "neutral" as a resource-provider for each tag he/she owns.
For example:
• I submitted an item resource of an RSS feed for Up-to-date Virtualization News
• I have attached 2 tags "virtualization" and "vm" to this item
• The system adds these tags to my profile with 50% value each, telling the community, that I am neutral when it comes to "virtualization" and "vm"
Rating Tags
Now, if I found another virtualization resource submitted by another person I can go ahead and value (vote) 100% "Highly Recommended" for that particular item's "virtulization" resource tag. The system will take that vote considering that I am only 50% "knowledgeable" in that particular resource and only increase that person's "virtualization" profile tag to about 62% (not 100%). The affect is small, because I am not that highly valued when it comes to "virtualization".
However, as time goes, and people start to highly value my "virtualization" resource (by voting on MY "virtualization" tag), then my "virtualization" resource tag will go up in value. So the next time I add a resource with that tag, it will be highly valued. Moreover, when I vote on an Item with a similar tag; that person's tag value will have a greater affect and so on...
This back-and-forth feedback by the community creates a complex web of value system.
But that's not all. The tags now act as a profile summary of a particular user. The tags hold the user's interest on different resources which brings us to the next important topic.
tag-suggest
In the works is a technology called tag-suggest™. It is a means to provide the user with suggested tags, resources, people, etc... appropriate for him/her.
When searching in dockex, tag-suggest™ will take into consideration and evaluate your list of tags, items, friends, etc... and their respective values in your profile
to suggest the best applicable tags, items or even other friends within the dockex community.
